Greater Greater Washington

Transit


How fast is your bus? WMATA maps bus speeds

As Bob Thomson reported on Sunday, WMATA has created maps showing the average speeds of buses across the region. These maps help illuminate where we could most help riders and also save money through strategic placement of bus lanes, queue jumpers, signal priority and more.

I'd been able to see early drafts of this map from WMATA officials. They have now finished the analysis and have posted it publicly.

The original maps showed slow buses in red, up to faster buses in green. However, the intermediate yellow bus lines were hard to see, and red to green is bad for color blind readers. Therefore, I've swapped the red and blue channels so that these maps show the bus speeds from green (fastest) through cyan to dark blue (slowest).

According to my color blindness simulator Photoshop plugin, color blind people should be able to distinguish all of these colors; the rivers and the fast buses might be similar colors, but they're easy to tell apart in other ways.

Closer to the core, the bus speeds are generally slower. That's a consequence partly of congestion, but also partly from bus stop density. Since there are more destinations and more riders in the central areas, there are more bus stops, and those stops have more riders, which take time to load and unload. DC could reduce some, but there will still be more and speeds will therefore be slower. It's where some lines are slower than their neighbors, sometimes much slower, that the difference becomes particularly useful.

Here are maps distinguishing the AM and PM peak times:

Based on this analysis, WMATA has identified a few key corridors ripe for improvement. These are areas where there are very slow buses coupled with high frequencies of buses. This is where a bus lane, queue jumper at an intersection, and/or signal priority could create the greatest time savings for the largest number of residents, and save the most money in operating costs as well.

This spreadsheet shows the full calculations. It includes the top ten corridors in each of the three jurisdictions for AM and PM peaks and all week, ranked by multiplying the number of buses per hour (for peak periods) or per week by the average speed over that link.

Local jurisdictions keep saying they are very strapped for cash and can't afford Metro extensions, light rail lines, big road widenings, new freeways, and more. Optimizing these high-priority corridors for efficient bus movement is the single most cost-effective, quick low-hanging fruit for improving mobility. Let's get on it, DOTs!

David Alpert is the Founder and Editor-in-Chief of Greater Greater Washington and Greater Greater Education. He worked as a Product Manager for Google for six years and has lived in the Boston, San Francisco, and New York metro areas in addition to Washington, DC. He loves the area which is, in many ways, greater than those others, and wants to see it become even greater. 

Comments

Add a comment »

How many times have I commented on the E/W Hwy J line buses? If you're building the Purple Line, build ridership by improving bus service between Silver Spring and Bethesda.

by Redline SOS on Jul 20, 2010 1:18 pm • linkreport

I am partially colorblind, and the green to blue looks great to me! Thank you!

by Paul C on Jul 20, 2010 1:48 pm • linkreport

Won't messing with signal timing will affect traffic generally? So it's hard to call this low-cost without considering the secondary impacts.

What are queue jumpers?

I like the idea of reducing stop density and evolving towards a culture where buses that are ahead of schedule can be flagged or add a dropoff in really bad weather, at the driver's discretion. But the normal service is stopping at fewer strategically placed stops. Stop density is way too high in many parts of the system.

Is it DOTs or DsOT?

by Ward 1 Guy on Jul 20, 2010 2:00 pm • linkreport

I'd just like to take this spot to say all the buses should be like the express buses. they are great. you can actually get somewhere in a reasonable amount of time. it is so frustrating riding a regular bus and having it stop at every damn block.

by RD on Jul 20, 2010 2:06 pm • linkreport

New Urbanism advocates on this blog have time and again asserted that TOD in places like Arlington has increased the density without increasing traffic.

This map proves they are wrong; the slowest areas are the ones near TOD (e.g., the R-B corridor and central DC). Contrast this with MacArthur Blvd in NW DC, Lee Highway west of Rosslyn, Washington Blvd west of Westover in Arlington, or even most of SE DC. All have less density, and in all of those places, the buses move faster, as shown on this map.

If you want relatively speedy bus transportation, you should not want there to be the towering, dense developments full of new residents--who will be either packing your Metro car or packing the highways your bus takes into work.

by JB on Jul 20, 2010 2:09 pm • linkreport

JB: See my note about bus stop density.

by David Alpert on Jul 20, 2010 2:13 pm • linkreport

David: I haven't measured how far apart the bus stops are in, say, western Arlington vs. inner DC--but my experience as a rider tells me they're not that different. E.g., the 3Y on Lee Hwy stops *very* frequently--and yet it moves pretty fast.

I give you some credit for acknowledging that the slowness in the inner corridor is partly due to increased density. (I believe it's the major cause.)

by JB on Jul 20, 2010 2:19 pm • linkreport

No, I'm saying it's the density of STOPS. The more people you have, the more often the buses stop and the longer it takes to load and unload. That's different from there being more traffic.

by David Alpert on Jul 20, 2010 2:20 pm • linkreport

@JB

Bus speeds are not a good measure of 'traffic.' Traffic volume is a good measure of traffic. As David notes, dense stop spacing will slow average bus speeds down. Likewise, slow speed limits will slow buses down.

by Alex B. on Jul 20, 2010 2:20 pm • linkreport

Hmm, looks like both of my busses are green. Not surprising, as my commute is suburb to suburb, connected by two rail legs in the core.

by Mike on Jul 20, 2010 2:35 pm • linkreport

One simple idea to increase bus speed would be a campaign to encourage, or better yet require, people to exit from the back door.

Particularly during rush hour, buses can sit for long periods while people exit from the front door which prevents people from loading onto the bus. If everyone exited from the back the time spent at bus stops could be cut in half.

Disabled passengers may have to exit through the front, but there is no reason why able bodied passengers should be taking up everyone's time (including their own) exiting through the front.

Has anyone considered this?

by Devoe on Jul 20, 2010 2:38 pm • linkreport

@ JB: I ride the 3's down Lee Highway a couple times a week and will be riding them every weekday for the next few months. East of Glebe Road, stops are very frequent and in some areas (namely the part of Lee Highway which isn't US 29 from Quincy to Lorcom) you can see one stop from another.

Re: the 3's running pretty fast, this is applicable to the 3Y and the 3A/B/E outside of rush hour. During rush hour, the 3's become a mess with lines to board at Rosslyn, frequent stops, and a high amount of bikers that use the route.

by Jason on Jul 20, 2010 2:39 pm • linkreport

They label 14th st. NW as 16th st. NW. Hopefully that was just a mistake in the rush to get public-ready versions out the door, but I'd hate to see them neglect it. I rode the S1/S2/S4 to work every day for several months, and they could use some serious decreases in stop density. The car traffic moves surprisingly quickly, with the synchronized lights, but the bus is constantly missing greens to drop off that one random person riding to Riggs Pl. or Corcoran St. who couldn't just walk an extra block.

by Jeff on Jul 20, 2010 2:59 pm • linkreport

@ Devoe

I think able-bodied passengers should exit from the rear of the bus, as that will reduce lead time (example from BART system). I normally ride my bike down 11th St. NW and notice that I move faster on my bike than the bus. This is due to the number of stops the bus makes.

One suggestion would be for bus stops that have higher density of people pay before they board the bus (I have seen this model for Bogota's RBT and streetcar models in Phoenix). It would have to be on an honor system of sorts, but I think the majority of people would honor the system if that means they can board the bus quicker.

There are primarily three factors (to me) that determine bus speed: people boarding/unboarding the bus, traffic light timing, and density of car traffic and pedestrian traffic.

by PC on Jul 20, 2010 3:03 pm • linkreport

On 16th. St. NW, I think they should make all buses express at morning and evening rush. Easy change to implement because S9 already has a good express route mapped out. As it stands, the S9's efficacy is significantly hampered by running in the same lane as S1/S2/S4, where it's constantly being forced to slow down and miss green lights by stopped "local route" buses. The main fix it would require, I think, is to build some bigger shelters at the express stops. They already overflow at rush hour as it is.

by Jeff on Jul 20, 2010 3:16 pm • linkreport

@Jeff

And what are those of us who don't live near one of the express stops on 16th supposed to do?

by MLD on Jul 20, 2010 3:18 pm • linkreport

@Ward1Guy:
Since no one else answered.
Queue Jump:
http://en.wikipedia.org/wiki/Queue_jump

by kidincredible on Jul 20, 2010 3:25 pm • linkreport

Not sure what the answer is, but I concur with the general sentiment about the S9 bus. I am supportive of express buses, but when I take it South from Irving to M St., there is usually not an appreciative difference in timing from the local buses. Traffic is just too heavy for passing and the S9 constantly gets stuck behind an S1/2/4 making an intermediate stop. Usually in the evening commute home, I just hop on a local bus, since it almost as fast and sometimes less crowded.

I think there are just too many stops south of U St. One of the stops, either Riggs or Corcoran, is literally a half block from the last one. Totally unnecessary. When I take the S9 to Silver Spring, its effectiveness seems to show more, however.

by 80p on Jul 20, 2010 3:54 pm • linkreport

RE: S9

The S9 isn't worth waiting for if you are going to or coming from south of Spring Rd I bet. But I choose it if it arrives at 16th & I first or is right behind another S bus, because it gets to 16th & Irving at least 5 minutes faster. It's only a 15-20 minute ride so 5 minutes is a lot.

by MLD on Jul 20, 2010 4:13 pm • linkreport

I worry that some GGW people want to use this data to argue for more less stops. We do need some more express buses during rush hour. But stops are important too.

After all, it isn't the density of stops. It is the density of stops when people USE that stop.

Perhaps express buses that have limited pick ups but unlimited drop off points would work. Pull up, only open the back door, and keep moving. After all, the delay in boarding is usually worse than the delay in exiting.

And again, start charging for bicycle racks on buses. Those really slow you down. Or prohibit them during rush hour.

by charlie on Jul 20, 2010 4:19 pm • linkreport

"I ride the 3's down Lee Highway a couple times a week and will be riding them every weekday for the next few months. East of Glebe Road, stops are very frequent and in some areas (namely the part of Lee Highway which isn't US 29 from Quincy to Lorcom) you can see one stop from another."

This is actually an understatement. These bus stops have to be some of the most ridiculous and redundant bus stops in the entire metro area. On this stretch of road you can see 3 bus stops in one direction. There is 10 story apartment complex there that has a bus stop on the west side of the building, at the middle of the building, and on the east side of the building. And it isn't that big a building. They should abolish 2 of these stops.

by db on Jul 20, 2010 11:44 pm • linkreport

@db If we're talking about the building between Thomas and Utah, keep in mind that those stops may have pre-dated the construction of that building.

In a similar vein, across the street from that building there's a complex with a stop at Thomas with the condo complex next door getting their own stop at Vermont. I think the stops at Thomas and Vermont could be eliminated with little opposition.

by Jason on Jul 21, 2010 8:43 am • linkreport

Why not run express buses on all routes.

Lets says a buses runs atleast 4 buses per hours lets say the 90 does. It could leave Anacostia at 10:00, 10:15, 10:30, and 10:45 lets take 2 of those trips (10:00 and the 10:30)and make them express and do this for all buses.

@ Devoe

I have been on buses long and short were you can not get to the back door because it is packed and I will then go to which ever door is closest.

If you are the only person getting off at a stop it is very hard to get to the back the door and usually the doors would open and close before you get there so you would have to yell or to get the driver to open the doors again.

How do you get to the door when you have to go through 30, 40 or 60 people who are not getting off.

The problem is the placement of the doors. We should using buses like the circulators or even buses with 4 doors which are used in other places that way when you're on the bus you arent that far from a door.

by kk on Jul 21, 2010 9:53 am • linkreport

You want to see a wonderful solution - Dedicated lanes (making buses a priority) A-la Bus Rapid Transit with a feeder system. As it now, I can out-bike most buses in DC and Baltimore - I'm freer, faster and fitter!
http://www.streetfilms.org/bus-rapid-transit-bogota/

by Greg Cantori on Jul 21, 2010 10:11 am • linkreport

@MLD- The S9 stops are not terribly far apart. Starting at 16th and P St going northbound, the gaps are 0.5 miles to U, .4 to Euclid, .4 to Irving, .2 to park, .4 to spring, .6 to Buchanan, .4 to Colorado, .7 to Missouri, .3 to Sheridan, 1.3 to Kalmia (along Walter Reed), and then you're basically at the metro. With the exceptions of Spring to Buchanan and Colorado to Missouri, all those gaps are .4 miles are less, meaning the furthest you'd have to walk along 16th, upon reaching it, to get to the nearest S9 stop, would be 1/5 of a mile. Further, those bigger gaps are largely bordered on at least one side by parkland. I'd grant that a stop at Walter Reed would be a good idea. Also, consider using 14th street buses for local service. They're only 2 blocks away, after all, and deliver people to similar downtown areas.

by Jeff on Jul 21, 2010 3:47 pm • linkreport

Add a Comment

Name: (will be displayed on the comments page)

Email: (must be your real address, but will be kept private)

URL: (optional, will be displayed)

Your comment:

By submitting a comment, you agree to abide by our comment policy.
Notify me of followup comments via email. (You can also subscribe without commenting.)
Save my name and email address on this computer so I don't have to enter it next time, and so I don't have to answer the anti-spam map challenge question in the future.

or